Homec4science
Diffusion Lenstool-HPC (configure) Browse

R1448/Benchmarks/GradientBenchmark/main.cpp
831158ed29e2configure

Last Change
Raw File

main.cpp
View Options

/**
* @file main.cpp
* @Author Christoph Schaaefer, EPFL (christophernstrerne.schaefer@epfl.ch)
* @date October 2016
* @brief Benchmark for gradhalo function
*/
#include <iostream>
#include <iomanip>
#include <string.h>
//#include <cuda_runtime.h>
#include <math.h>
#include <sys/time.h>
#include <fstream>
//
//
#include <mm_malloc.h>
//#define __WITH_LENSTOOL 0
//
#ifdef __WITH_LENSTOOL
#warning "linking with libtool..."
#include<fonction.h>
#include<constant.h>
#include<dimension.h>
#include<structure.h>
//#include <liblenstool>
#endif
//#include "../../../Projects/lenstool-6.8.1/include/structure.h"
//#include "structure.h"
#include "structure_hpc.hpp"
#include "timer.h"
#include "gradient.hpp"
#include "gradient_avx.hpp"
#include "gradient_avx512f.hpp"
#include "setup.hpp"
#include <type.h>
//
#define NN 100000
//
//
//
#ifdef __WITH_LENSTOOL
struct g_mode M;
struct g_pot P[NPOTFILE];
struct g_pixel imFrame, wFrame, ps, PSF;
struct g_cube cubeFrame;
struct g_dyn Dy; // //TV
struct g_source S;
struct g_image I;
struct g_grille G;
struct g_msgrid H; // multi-scale grid
struct g_frame F;
struct g_large L;
struct g_cosmo C;
struct g_cline CL;
struct g_observ O;
struct pot lens[NLMAX];
struct pot lmin[NLMAX], lmax[NLMAX], prec[NLMAX];
struct g_cosmo clmin, clmax; /*cosmological limits*/
struct galaxie smin[NFMAX], smax[NFMAX]; // limits on source parameters
struct ipot ip;
struct MCarlo mc;
struct vfield vf;
struct vfield vfmin,vfmax; // limits on velocity field parameters
struct cline cl[NIMAX];
lensdata *lens_table;
int block[NLMAX][NPAMAX]; /*switch for the lens optimisation*/
int cblock[NPAMAX]; /*switch for the cosmological optimisation*/
int sblock[NFMAX][NPAMAX]; /*switch for the source parameters*/
int vfblock[NPAMAX]; /*switch for the velocity field parameters*/
double excu[NLMAX][NPAMAX];
double excd[NLMAX][NPAMAX];
/* supplments tableaux de valeurs pour fonctions g pour Einasto
* * Ce sont trois variables globales qu'on pourra utiliser dans toutes les fonctions du projet
* */
#define CMAX 20
#define LMAX 80
float Tab1[LMAX][CMAX];
float Tab2[LMAX][CMAX];
float Tab3[LMAX][CMAX];
int nrline, ntline, flagr, flagt;
long int narclet;
struct point gimage[NGGMAX][NGGMAX], gsource_global[NGGMAX][NGGMAX];
struct biline radial[NMAX], tangent[NMAX];
struct galaxie arclet[NAMAX], source[NFMAX], image[NFMAX][NIMAX];
struct galaxie cimage[NFMAX];
struct pointgal gianti[NPMAX][NIMAX];
struct point SC;
double elix;
double alpha_e;
double *v_xx;
double *v_yy;
double **map_p;
double **tmp_p;
double **map_axx;
double **map_ayy;
#endif
//struct pot lens_ref[NLMAX];
//
//#include <iitnotify.h>
int main()
{
double t0, t1, t2, t3;
//
//Variable creation
//
point image;
//
//Initialisation
//
int nlenses;
type_t x, y;
type_t sol_grad_x, sol_grad_y;
point grad; // store the result
//pot* lens;
#ifdef __WITH_LENSTOOL
setup_jauzac_LT(&nlenses, &image.x, &image.y, &sol_grad_x, &sol_grad_y);
//
struct point grad_lt, Grad;
//printf("Number lenses = %d, b0 = %f\n", nlenses, lens[0].b0);
//
t0 = -myseconds();
for (int ii = 0; ii < NN; ++ii)
{
grad_lt.x = grad_lt.y = 0.;
for (long int jj = 0; jj < nlenses; ++jj)
{
//printf("%f\n", lens[ii].b0);
Grad = e_grad_pot(&image, jj);
//
grad_lt.x += Grad.x;
grad_lt.y += Grad.y;
//printf("%f %f\n", grad_lt.x, grad_lt.y);
}
}
t0 += myseconds();
#endif
//
//
//setup_jauzac(Potential** lens, int* nlenses, double* x, double* y, double* sol_grad_x, double* sol_grad_y)
//
//
Potential* lens_aos;
setup_jauzac(&lens_aos, &nlenses, &image.x, &image.y, &sol_grad_x, &sol_grad_y);
t1 = -myseconds();
for (int ii = 0; ii < NN; ++ii)
grad = module_potentialDerivatives_totalGradient(nlenses, &image, lens_aos);
t1 += myseconds();
//printf("---> grad = %f %f\n", grad.x, grad.y);
//
// Setting up the AOS potential
//
point image_soa;
int nlenses_soa;
double x_soa, y_soa;
type_t sol_grad_x_soa, sol_grad_y_soa;
//
Potential_SOA lens_soa;
setup_jauzac_SOA(&lens_soa, &nlenses_soa, &image_soa.x, &image_soa.y, &sol_grad_x_soa, &sol_grad_y_soa);
//
std::cout << "Benchmark for Gradient Calculation using " << nlenses_soa << " lenses with type " << lens_soa.type[0] << ", image: " << image_soa.x << " " << image_soa.y << std::endl;
//
// AVX version
//
#if 0
point grad_soa_novec;
double t21 = -myseconds();
for (int ii = 0; ii < NN; ++ii)
{
//grad_soa = module_potentialDerivatives_totalGradient_SOA_AVX512(&image_soa, &lens_soa, nlenses_soa);
//grad_soa = module_potentialDerivatives_totalGradient_8_SOA_AVX(&image_soa, &lens_soa, nlenses_soa);
//
//grad_soa_avx = module_potentialDerivatives_totalGradient_81_SOA_AVX(&image_soa, &lens_soa, 0, nlenses_soa);
grad_soa_novec = module_potentialDerivatives_totalGradient_SOA_novec(&image_soa, &lens_soa, nlenses_soa);
//grad_soa = module_potentialDerivatives_totalGradient_SOA(&image_soa, &lens_soa, 0, nlenses_soa);
}
t21 += myseconds();
std::cout << "1" << std::endl;
#endif
//
// no vectorization
//
#if 1
point grad_soa; // store the result
t2 = -myseconds();
for (int ii = 0; ii < NN; ++ii)
{
//grad_soa = module_potentialDerivatives_totalGradient_SOA_AVX512(&image_soa, &lens_soa, nlenses_soa);
//grad_soa = module_potentialDerivatives_totalGradient_8_SOA_AVX(&image_soa, &lens_soa, nlenses_soa);
//
//grad_soa_avx = module_potentialDerivatives_totalGradient_81_SOA_AVX(&image_soa, &lens_soa, 0, nlenses_soa);
grad_soa = module_potentialDerivatives_totalGradient_SOA(&image_soa, &lens_soa, nlenses_soa);
//grad_soa = module_potentialDerivatives_totalGradient_SOA(&image_soa, &lens_soa, 0, nlenses_soa);
}
t2 += myseconds();
#endif
//__SSC_MARK(0x222);
//
// autovectorized version
//
point grad_soa_avx;
t3 = -myseconds();
for (int ii = 0; ii < NN; ++ii)
{
//grad_soa_avx = module_potentialDerivatives_totalGradient_SOA_AVX512(&image_soa, &lens_soa, nlenses_soa);
// grad_soa = module_potentialDerivatives_totalGradient_81_SOA_AVX(&image_soa, &lens_soa, nlenses_soa);
// //grad_soa = module_potentialDerivatives_totalGradient_81_SOA_AVX(&image_soa, &lens_soa, nlenses_soa);
//grad_soa = module_potentialDerivatives_totalGradient_81_SOA(&image_soa, &lens_soa, 0, nlenses_soa);
//grad_soa = module_potentialDerivatives_totalGradient_8_SOA(&image_soa, &lens_soa, 0, nlenses_soa);
#ifdef _double
grad_soa_avx = module_potentialDerivatives_totalGradient_SOA_AVX(&image_soa, &lens_soa, nlenses_soa);
#endif
}
t3 += myseconds();
//
#ifdef _double
std::cout << " Double calculation = " << std::endl;
#else
std::cout << " Float calculation = " << std::endl;
#endif
std::cout << " ref sol = " << std::setprecision(15) << sol_grad_x << " " << std::setprecision(15) << sol_grad_y << std::endl;
//
#ifdef __WITH_LENSTOOL
std::cout << " Lenstool sol = " << std::setprecision(15) << grad_lt.x << " " << std::setprecision(15) << grad_lt.y << ", time = " << t0 << " s." << std::endl;
#endif
//
std::cout << " grad = " << std::setprecision(15) << grad.x << " " << std::setprecision(15) << grad.y << ", time = " << t1 << " s., speedup = " << (double) t0/t1 << std::endl;
//
//std::cout << " grad novec = " << std::setprecision(15) << grad.x << " " << std::setprecision(15) << grad.y << ", time = " << t21 << " s., speedup = " << (double) t0/t21 << std::endl;
//
std::cout << " grad SIMD = " << std::setprecision(15) << grad_soa.x << " " << std::setprecision(15) << grad_soa.y << ", time = " << t2 << " s., speedup = " << (double) t0/t2 << std::endl;
//
std::cout << " grad handcoded = " << std::setprecision(15) << grad_soa_avx.x << " " << std::setprecision(15) << grad_soa_avx.y << ", time = " << t3 << " s. speedup = " << (double) t0/t3 << std::endl;
//
}
c4science · Help